Lyria 3 by Google. The musical face of Gemini! – AI News – #4 February 2026

From now on, the Gemini app doesn’t just seamlessly generate text, images, and video; it becomes a comprehensive, multimedia assistant. The latest Lyria 3 model, developed by engineers at Google DeepMind, is currently rolling out in beta for users 18 and older. It allows for the instant creation of short, catchy, and technically advanced tracks ready to be shared on social media.

The most important capabilities and features of the new model include:

The tool is already available in several languages, including English, German, Spanish, French, Japanese, and Portuguese, with Google planning further expansion.

In parallel, the powerful capabilities of the Lyria 3 model are being handed to video creators as part of the global expansion of the Dream Track feature on YouTube. This option, which until now was tested primarily in the United States, makes it much easier and faster to create personalized, professional-sounding backing tracks for short-form videos (YouTube Shorts). This allows creators to generate unique verses or background music that perfectly matches the dynamics of their content, taking viewer engagement to a whole new level.

The development of such powerful generative tools in the music industry must go hand in hand with responsibility and respect for intellectual property. Google emphasizes that Lyria 3 was designed exclusively as a tool for original expression, not a machine for directly copying and mimicking specific artists. If you include a well-known musician’s name in your prompt, the algorithm will treat it only as a broad, loose stylistic inspiration, and built-in filters are in place to prevent plagiarism.

To maintain full transparency and effectively combat abuse, specific preventive measures have been introduced:

Despite the technological awe and obvious benefits for internet creators, we must face the truth: the growing power of artificial intelligence in music also has a very dark side. We are talking primarily about the very real phenomenon of flooding major streaming platforms (such as Spotify, Apple Music, or Tidal) with hundreds of thousands of tracks generated massively and algorithmically.

For niche, independent artists, this represents a direct, sometimes even existential, threat. Real musicians dedicate enormous effort, spend hours in rehearsals, pay for expensive studio time, invest in mixing and mastering, and finally incur significant financial costs to promote their work. Unfortunately, it is now incredibly easy for a person with zero musical skills to generate dozens of songs a day and sign as the “creator” under something that was made—to put it bluntly—shamelessly and without the slightest effort.

The consequences are brutal:

While the democratization of the creative process sounds beautiful, the current situation raises very serious and fully justified questions about the future value of human creativity. It also highlights the burning need to implement new, global regulations in the music market.

Do you want to stay up to date with the most important news from the world of AI and SEO? Algorithms and technologies change in the blink of an eye today, and we keep our finger on the pulse daily so you don’t have to waste your time doing it. Sign up for the free Delante newsletter to regularly receive expert analyses, the latest industry news, and practical tips straight to your inbox. Don’t let your competition get ahead, and build your digital advantage today!

Source of information: https://deepmind.google/models/lyria/